Image Credits: Raimundas/shutterstock.com
For decades, we relied on silicon as the semiconductor for our computer chips. But now, working at nanometer scales, it looks like physical limitations may end the current methods to include more and more processing power onto each individual chip.
Many companies are making billion-dollar investments to continue scaling down semiconductor technology. The pressures of big data and cloud computing are pushing the limits of the current semiconductor technology in terms of bandwidth, memory, processing speed, and device power consumption.
Today’s state-of-the-art silicon chips are engineered at the 22- and 14-nanometer scale. Research is underway to take that down to 10-nanometer scale in the next several years.
However at some point we will be required to switch over to a different kind of technology to handle the electronics challenges we will be facing 100 years from now.
Graphene is a single sheet of carbon atoms arranged in a two-dimensional honeycomb lattice, with one atom at each vertex.
In a recent study, researchers revealed that lattice structure is capable of forming wrinkles, giving it potential as a semiconductor.
These wrinkles can limit the motion of electrons to one direction, establishing a junction-like framework that alternates from a zero-gap conductor to semiconductor back to a zero-gap conductor. The researchers used a scanning tunneling microscope to control the development of wrinkles, showing that graphene semiconductors could indeed be manufactured.
Graphene offers at least two major advantages over the conventional use of silicon as a semiconductor. As the size of a carbon atom is around 0.3 nanometers, graphene could pack a lot more processing power into a much smaller space. Also, insufficient heat dissipation is a major problem with silicon – a problem graphene does not have.
Unlike graphene, carbon nanotubes are natural semiconductors without any added molecular manipulation.
However, scientists have had difficulty making transistors with enough carbon nanotubes to hold a good amount of current. Carbon nanotube production methods also result in a blend of both metallic and semiconducting forms, which is unwanted. Furthermore, the number of nanotubes must be the same from transistor to transistor to be able to obtain any energy and performance advantages.
A team of Stanford engineers has been working to solve these problems, and last year the team announced that they can produce carbon nanotube transistors with current densities and other qualities comparable to that of silicon transistors. The team also said it was capable of assembling carbon nanotube transistors atop silicon circuits.
Even after solving these problems, the Stanford team’s nanotube transistors have current-moving channels that are 400 nanometers long, much longer than state-of-the-art conventional devices.
While carbon nanotube semiconductors do have a long way to go to become viable, one or two engineering breakthroughs could make them the industry standard of future electronics.
Despite all of the research dedicated to finding new semiconductor materials, the future of computing could still be based on silicon. Computer scientists largely agree that quantum computing is the future of computing, and some quantum computing research has found success with silicon.
A quantum computer holds and processes information based on the laws of quantum mechanics, leveraging those laws to unprecedented effect. For instance, a quantum bit, or qubit, can be put into a state where it is both a 0 and a 1. When a quantum computer performs logic operations that include this third, previously-unavailable option, it allows work to be accomplished at very fast speeds.
Image Credits: welcomia/shutterstock.com
Researchers are currently investigating numerous ways to create a qubit out of silicon.
One method uses a phenomenon known as carrier freeze-out. Occurring at a temperature so cold, most conventional silicon devices stop working, freeze-out produces a clutch of electrically neutral, isolated atoms that are all set in place - a collection of naturally-stable quantum frameworks to hold information.
Another approach makes qubits from single electrons trapped inside nanoscale constructions called quantum dots. Inside a quantum dot, electrons can be restricted so snugly that they’re compelled to occupy discrete energy amounts in the same way they would around an atom. The spin state of a trapped electron can be used as the cornerstone for a qubit.
What Does the Future Hold?
The future of electronics will undoubtedly be driven by the future of semiconductors, but what will those electronics even look like?
Many experts say that technology will increasingly become part of our very own biology. In addition to being able to perform nanoscale surgery, humans in the 22nd century will likely be augmenting their bodies with technology.
Experts are also predicting the development of immersive and highly-realistic virtual reality technology. Possible with the kind of processing power predicted for next-generation semiconductors, this kind of technology would revolutionize communications, entertainment, and countless other industries.
Sources and Further Reading